Browse Datasets

AfriQA

Cross-lingual Question Answering for 10 African Languages

question-answering cross-lingual reading-comprehension 12,000+ QA pairs CC-BY-4.0-NC
Access Dataset Read Paper

Description

First cross-lingual Open-Retrieval QA dataset for African languages. 12,000+ examples where answers might be in different language than question.

Quick Start

Python 
from datasets import load_dataset

dataset = load_dataset("masakhane/afriqa", "bem", split="train")
print(dataset['question'][:3], dataset['answers'][:3])

Project Ideas

Retrieval-based QA

Build retrieval-based QA for Bemba using English documents

Advanced

mBERT Fine-tuning

Fine-tune mBERT for answer span extraction in Yoruba

Intermediate

Ethical Considerations

  • Ensures African languages are first-class citizens in QA